Faster R-CNN与Region Proposal Networks的实时目标检测

需积分: 9 105 浏览量更新于2024-09-11 收藏 2.26MB PDF 举报

"Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks" 在计算机视觉领域，Fast R-CNN（快速区域卷积网络）是一种重要的目标检测算法，它由Shaoqing Ren、Kaiming He、Ross Girshick和Jian Sun等人在2015年提出。Fast R-CNN在当时的对象检测技术中引入了显著的改进，显著提升了检测速度，同时保持了较高的精度。 Fast R-CNN的核心改进在于解决了早期目标检测系统中的一个关键瓶颈——区域提案（region proposal）的计算效率问题。传统的检测网络如RCNN（区域卷积网络）依赖于像Selective Search这样的外部算法来生成可能包含对象的候选区域，然后对每个区域进行分类和定位。这种方法虽然效果不错，但处理速度较慢，因为每个候选区域都需要单独通过完整的CNN（卷积神经网络）进行处理。 Fast R-CNN的创新之处在于它将区域提案步骤与分类和定位步骤整合到了同一个网络中。具体来说，它首先使用预训练的CNN对整个图像进行一次前向传播，生成共享的特征图，然后在这个特征图上对每个候选区域进行滑动窗口操作，提取固定大小的区域特征进行分类和边界框回归。这样就避免了对每个候选区域重复计算CNN，大大提高了效率。然而，即便如此，区域提案仍然是一个耗时的过程。为了进一步提升速度，研究人员在Fast R-CNN的基础上提出了Faster R-CNN，引入了Region Proposal Network（RPN）。RPN是一个全卷积网络，它直接在共享的全图卷积特征上预测每个位置的对象边界框和“对象性”得分。对象性得分用于衡量该位置是否存在对象的可能性。RPN经过端到端的训练，能够生成高质量的区域提案，这些提案随后被Fast R-CNN用于检测。 Faster R-CNN的这种设计使得区域提案和检测可以在同一网络架构中交替优化，并且共享卷积层，从而极大地提高了整个系统的运行效率。例如，当使用深度较大的VGG-16模型时，Faster R-CNN能够在GPU上实现每秒5帧的检测速率，同时保持了最先进的检测性能。 Fast R-CNN和Faster R-CNN的出现是目标检测领域的里程碑，它们不仅提高了检测速度，而且推动了后续的YOLO（You Only Look Once）、SSD（Single Shot MultiBox Detector）等实时目标检测算法的发展，对现代计算机视觉应用产生了深远影响。

Faster R-CNN: Towards Real-Time Object Detection

with Region Proposal Networks

Shaoqing Ren

∗

Kaiming He Ross Girshick Jian Sun

Microsoft Research

{v-shren, kahe, rbg, jiansun}@microsoft.com

Abstract

State-of-the-art object detection networks depend on region proposal algorithms

to hypothesize object locations. Advances like SPPnet [7] and Fast R-CNN [5]

have reduced the running time of these detection networks, exposing region pro-

posal computation as a bottleneck. In this work, we introduce a Region Proposal

Network (RPN) that shares full-image convolutional features with the detection

network, thus enabling nearly cost-free region proposals. An RPN is a fully-

convolutional network that simultaneously predicts object bounds and objectness

scores at each position. RPNs are trained end-to-end to generate high-quality

region proposals, which are used by Fast R-CNN for detection. With a simple

alternating optimization, RPN and Fast R-CNN can be trained to share convolu-

tional features. For the very deep VGG-16 model [18], our detection system has a

frame rate of 5fps (including all steps) on a GPU, while achieving state-of-the-art

object detection accuracy on PASCAL VOC 2007 (73.2% mAP) and 2012 (70.4%

mAP) using 300 proposals per image. The code will be released.

1 Introduction

Recent advances in object detection are driven by the success of region proposal methods (e.g., [21])

and region-based convolutional neural networks (R-CNNs) [6]. Although region-based CNNs were

computationally expensive as originally developed in [6], their cost has been drastically reduced

thanks to sharing convolutions across proposals [7, 5]. The latest incarnation, Fast R-CNN [5],

achieves near real-time rates using very deep networks [18], when ignoring the time spent on region

proposals. Now, proposals are the computational bottleneck in state-of-the-art detection systems.

Region proposal methods typically rely on inexpensive features and economical inference schemes.

Selective Search (SS) [21], one of the most popular methods, greedily merges superpixels based

on engineered low-level features. Yet when compared to efﬁcient detection networks [5], Selective

Search is an order of magnitude slower, at 2s per image in a CPU implementation. EdgeBoxes

[23] currently provides the best tradeoff between proposal quality and speed, at 0.2s per image.

Nevertheless, the region proposal step still consumes as much running time as the detection network.

One may note that fast region-based CNNs take advantage of GPUs, while the region proposal meth-

ods used in research are implemented on the CPU, making such runtime comparisons inequitable.

An obvious way to accelerate proposal computation is to re-implement it for the GPU. This may be

an effective engineering solution, but re-implementation ignores the down-stream detection network

and therefore misses important opportunities for sharing computation.

In this paper, we show that an algorithmic change—computing proposals with a deep net—leads

to an elegant and effective solution, where proposal computation is nearly cost-free given the de-

tection network’s computation. To this end, we introduce novel Region Proposal Networks (RPNs)

∗

Shaoqing Ren is with the University of Science and Technology of China. This work was done when he

was an intern at Microsoft Research.

arXiv:1506.01497v1 [cs.CV] 4 Jun 2015

下载后可阅读完整内容，剩余9页未读，立即下载

人工智能的弱者

粉丝: 8
资源: 24

Faster R-CNN与Region Proposal Networks的实时目标检测

Fast RCNN深度网络

faster_rcnn_models训练模型的下载地址

代码FastRcnn

我需要在组会讲解R-CNN、Fast R-CNN、Faster R-CNN、Mask R-CNN，请为我提出一些思路和具体的讲解内容

基于候选区域的深度学习目标检测算法R-CNN，Fast R-CNN，Faster R-CNN

在一个数据集上，测试 R-CNN、Fast R-CNN、Faster R-CNN算法，并对比优缺点

Faster R-CNN算法与Fast R-CNN算法的区别详细介绍

Fast R-CNN

R-CNN、Fast R-CNN和Faster R-CNN分别是什么？

在一个数据集上，测试Fast R-CNN、Faster R-CNN、YOLO算法

最新资源